Digital Marketing

How to utilize attribution models in Google Analytics

Working within e-commerce, you have most likely come in contact with the concept of attribution models. Attribution is a hot topic within e-commerce and digital marketing since the customer journey is becoming more and more complex with several channels intertwined. When analyzing your data sources, you can get a deeper understanding of your visitors’ touchpoints before they convert into customers.

Attribution models in Google Analytics can help you measure and analyze the value of the different channels, but it’s not always easy to understand. Here we will help you analyze and compare the attribution models in your Analytics account.

What is an attribution model?

When a visitor has completed a purchase, you can enter Google Analytics and see which traffic source has generated the sale. The data can be deceptive already at this point since the customer rarely converts on the first visit. On the contrary – a conversion takes often place after several visits on the website and via multiple digital channels.

Google Analytics reporting is most often set on the “Last Non-Direct Click” model. The traffic source/channel/campaign/ad/link that was the last interaction (excluding possible direct visit) gets the whole conversion value attributed. Analyzing exactly which channels that cooperate and the path to conversion will help you to get a better understanding of how the different marketing channels work together to generate sales.

In Google Analytics under Multi-Channel Funnels → Overview you get a summary of the visitors’ click paths to conversion.

 

Google Analytics attribution definitions

You find the different Google Analytics attribution models under Conversions → Multi-Channel Funnels → Model Comparison Tool

Last Interaction
The last touchpoint will be attributed as the only contributing factor to why the conversion took place.

Last Non-Direct Click
This model assigns the last known data source 100% of the conversion value. This attribution model is set as a standard in Analytics, unless you have actively changed this.

First Interaction
In this case, 100% of the conversion value is assigned to the first traffic source.

Linear
All visitor contact points are assigned the equal value regardless of how many visits/channels brought the visitor to the website.

Position Based
Position based modeling entails that the first and last interaction are assigned the majority of the conversion value.

Time Decay
With this attribution model, the contact points closest to purchase are assigned the majority of the conversion value.

 

To choose the right attribution model

The reason why there are so many attribution models is that there is not one single model that fits all. In general, you can say that models that are more generous towards channels with higher conversion rates, fit better in a profit-focused business. A model that is more generous towards channels with assisted conversions, fit better for growth-oriented businesses.

Customer journeys can vary a lot. As an example, a customer can visit your website from a referral link, an ad on Google and an organic search result, and in the end convert through a visit directly in the web browser. This means there are four contact points. When you change the attribution model, the channel value will change. But which channel that is most fair, is up for you to decide. When implementing an attribution model adapted to your business, you will get much better insights into your marketing effects since you can see how different campaigns impact the sales in total.

 

Other insights

Besides analyzing the channels your customer interacts with, there are even more parameters to consider. Under the section Multi-Channel Funnels you can also find data on path length, that is the number of contact points your visitor has had before converting and also the duration from interaction to purchase.